Endangered Species Research 35:125

Home , Santolina

Vol. 35: 125–139, 2018 ENDANGERED SPECIES RESEARCH Published March 15 https://doi.org/10.3354/esr00878 Endang Species Res

OPENPEN ACCESSCCESS

Representation and protection of threatened biodiversity by the largest Spanish regional network of protected areas

David Rodríguez-Rodríguez1,2,*, Javier Martínez-Vega1

1Institute of Economy, Geography and Demography, Spanish National Research Council (IEGD-CSIC), Associated unit GEOLAB, 28037 Madrid, Spain 2Andalucía Tech, European Topic Centre, University of Malaga, 29010 Málaga, Spain

ABSTRACT: The main global strategy to stop biodiversity loss is the designation of protected areas (PAs). Spain is a highly biodiverse country. It has one of the world's greatest terrestrial PA coverages. However, the status of its biodiversity is delicate as a result of serious pressures, and some important areas for biodiversity are outside PAs. We used official census data to spatially assess how 71 habitats of community interest (HCIs), 126 regionally threatened flora, fauna and fungi species and subspecies (RTSs), and 33 globally threatened species or subspecies (GTSs) are represented in a network of 404 PAs in Andalusia, a region rich in biodiversity in southern Spain. We also assessed the legal and managerial protection afforded to these threatened habitats and species by those PAs. The Andalusian PA network expands across one-third of the region’s territory and includes the threatened species’ richest areas. However, it only covers 57% of the area of occupancy of RTSs, 81% of the regional area of occupancy of GTSs, and 53% of the extent of HCIs. Over 61% of the regional PA network area is assigned more than 1 PA designation category, although cumulative legal protection is marginally related to RTS richness and unrelated to GTS richness. RTSs and especially GTSs occupy the most relative area in Ramsar sites (i.e. wetlands of international importance), although these are of relatively minor importance for threatened habitats. Wetlands and agricultural areas are the broad ecosystem types showing the greatest numbers of RTSs and GTSs. Seven GTSs were not included in the Andalusian Register of Threatened Spe- cies. One hundred and eleven unprotected Areas of High Importance for Threatened Species (AHITSs) and one Area of High Importance for Threatened Biodiversity (AHITB) were identified. Those species and sites are good candidates for a targeted expansion of legal protection of bio - diversity in the region.

KEY WORDS: Endangered species · Endangered habitats · Land use-land cover · Gap analysis · Reserves · Andalusia

INTRODUCTION change and the introduction of alien invasive species (Gaston et al. 2008, Butchart et al. 2010, WWF 2016). Biodiversity is defined as the variability among liv- The main global strategy to stop global biodiversity ing organisms, including diversity within species, be - loss is the designation of protected areas (PAs) (Dud- tween species and of ecosystems (CBD 1992). Global ley 2008, Gaston et al. 2008). In the European Union biodiversity is seriously declining due to human (EU), the Habitats Directive (EEC 1992) created a activities which result in habitat destruction, frag- wide ecological network of PAs to protect sensitive mentation and pollution, direct persecution, climate species and habitats in Europe: the Natura 2000 net-

© The authors 2018. Open Access under Creative Commons by *Corresponding author: [email protected] Attribution Licence. Use, distribution and reproduction are un - restricted. Authors and original publication must be credited. Publisher: Inter-Research · www.int-res.com 126 Endang Species Res 35: 125–139, 2018

work. It distinguishes 2 types of PAs: Sites of Com- the EU (Andalusian Government 2016a). However, munity Importance (SCIs)/Special Areas of Conser- most regional PA designations occurred in the late vation (SACs), for the conservation of natural and 1980s on the basis of limited ecological knowledge, semi-natural habitats and wild species of flora and when Andalusia was transferred the administrative fauna, except bird species; and Special Protection competencies for designation and management of Areas (SPAs), for the conservation of wild bird spe- PAs and passed its own law on PAs (Andalusian Gov- cies, according to the EU Birds Directive (EEC 1979). ernment 1989). Yet the first PA in the region was des- The Natura 2000 network includes >27 000 sites that ignated as early as 1969 (i.e. Doñana National Park). span across 18% of the terrestrial territory of the EU Since those days, PA designation criteria and ecolog- (EC 2016). Despite the continuous growth in PA cov- ical knowledge have evolved (Múgica et al. 2002), erage worldwide, concerns are expressed that impor- making it advisable to assess whether the regional tant areas for biodiversity are not included in PA net- PA network has grown in an ecologically consistent works, that PA effectiveness evaluations still need to manner and is currently fit for the purpose of con- be generalised (Juffe-Bignoli et al. 2014, Bhola et al. serving threatened biodiversity, as national and 2016) and that protection afforded to existing PAs international regulations require (Spanish Govern- might not be sufficient to avert biodiversity loss ment 2007, CBD 2010). (Craigie et al. 2010, Laurance et al. 2012). Therefore, this study seeks to (1) spatially assess Spain is a highly biodiverse country in the Euro- whether threatened biodiversity, including species, pean context (Múgica et al. 2010). It is located in the subspecies and habitats, is adequately represented Mediterranean basin, one of the global biodiversity in the regional PA network; (2) ascertain the regional hotspots critical for biodiversity conservation (Myers distribution of threatened biodiversity by broad eco- et al. 2000). Moreover, southeastern Spain is consid- system types; (3) estimate the legal and managerial ered one of the top 10 fine-scale biodiversity hotspots protection afforded to threatened biodiversity; and around the Mediterranean basin in terms of plant (4) spatially analyse whether areas rich in threatened species richness and endemicity (Médail & Quézel biodiversity exist outside current PAs, with a view to 1999). Accordingly, Spain's terrestrial PA network proposing sustainable territorial planning and man- covers more than 27% of its land and freshwater ter- agement recommendations. ritory (Múgica et al. 2014), largely exceeding international PA coverage targets at 17% by 2020 (CBD 2010). However, the status of Spanish biodiversity, METHODS especially of coastal biodiversity, is delicate, mainly due to extensive habitat destruction and degradation Study area in recent decades (Estévez et al. 2016), and important areas for biodiversity are outside PAs (Muñoz- Andalusia is the second biggest autonomous Rodríguez et al. 2016, Zamora-Marín et al. 2016). In region in Spain, covering >87 000 km2, or 17% of the addition, restoration plans for threatened species are country’s territory. It is entirely in the Mediterranean scarce and not always implemented (Jiménez 2012). biogeographical region (EEA 2012), where nearly Finally, regular, systematic and comprehensive eval- half of the species and more than half of the habitats uation of the effectiveness of the country’s PAs is re- included in the Habitats Directive are found (Barredo stricted to the National Park network (Rodríguez- et al. 2016), and where habitat conversion exceeds Rodríguez et al. 2015a) and the periodic (though of habitat protection by a ratio of 8:1 (Hoekstra et al. improvable completeness and accuracy) reporting on 2005). the status of species and habitats under Habitats Andalusia was chosen for this study for being a bio- Directive obligations (EEC 1992). diversity-rich region subject to numerous threats Spain is a highly decentralised country, with 17 (Andalusian Government 2014, 2015). It also has the regional governments. The Spanish government is most comprehensive, consistent and updated official responsible for making and passing basic legislation repository of data on biodiversity of conservation on nature conservation that all the regions must importance of all the Spanish regions. Moreover, in abide by. However, most nature conservation poli- 2013 almost 19% of Andalusian territory was cov- cies, including the legal protection and management ered by regional PA designation categories, reaching of species and PAs, are regional competencies in 30% if Natura 2000 sites were included, making it Spain. The Andalusian network of PAs is the broad- the largest regional PA network in Spain in absolute est and most numerous sub-national PA network in terms (Múgica et al. 2014; our Fig. 1). Finally, biodi- Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 127

Endangered, Endangered, or Vulnerable (globally threatened species or GTSs) were selected for our analysis of threatened species (IUCN 2017). Regional registers of threatened species are official lists that must abide by the Spanish Register of Threatened Species (Spanish Government 2007, 2011), which classifies species according to their degree of threat at the country scale based on specific scientific criteria on population trends, areas of occupancy, or population viability analyses (Spanish Gov- ernment 2017). Regional registers must include the species in the Spanish Register that inhabit the region of concern with at least the same threat category as in the Spanish Register or higher, as assessed at the regional scale. The inclusion of a taxon in one Fig. 1. Andalusia and its protected area network in the admin- of these registers implies a legal obligation for the istrative regional map of Spain (except the Canary Islands respective regional government to conduct periodic Region) censuses and conservation actions, depending on the category of threat. versity management is a regional competency in Spatial data for the 71 HCIs identified in the region Spain. Thus, from a practical, administrative point of (Andalusian Government 2015) were downloaded view, regional analyses are the most meaningful from the regional ministry’s environmental informa- scale for proposing biodiversity conservation meas- tion website (Andalusian Government 2016b). For ures in the country. habitats, we considered their regional extent, a globally recommended criterion for assessing the conservation status of habitats (Bland et al. 2017). Both the Definitions, assumptions and data sources complete species layer and the HCIs layer included data updated in 2015. In this study, we considered ‘threatened biodiver- We defined regional Areas of High Importance for sity’ to be (1) threatened species and subspecies, and Threatened Species (AHITSs) as those unprotected (2) habitats of community interest (HCIs). For linguis- areas ≥100 ha in which at least 1 GTS or 3 RTSs tic simplicity, we will refer to ‘species’ and ‘subspecies’ occur. Areas of High Importance for Threatened Bio- simply as ‘species’. According to the Habitats Direc- diversity (AHITBs) were defined as AHITSs that spa- tive, HCIs are those entirely natural or semi-natural tially coincide with HCIs. New possible PAs in the habitats that, within the territory of EU member states region were identified by selecting areas that qualify (1) are endangered, or (2) have a small natural range, as AHITSs or AHITBs. To assess ‘protection’ of PAs, or (3) are representative of the biogeograhical regions we followed the definition by Rodríguez-Rodríguez present in the territory of the EU (EEC 1992). et al. (2016), who discriminated between ‘legal pro- We were provided comprehensive official vector tection’ and ‘management effort’. distribution data for terrestrial and freshwater spe- The digital boundaries of PAs were obtained from cies of flora, fauna and fungi for which official moni- the Andalusian Ministry of Environment’s digital re - toring had been done in the region (Andalusian Gov- pository (Andalusian Government 2017c). They in - ernment 2017a). These data were provided by the cluded all national and international PA designation Regional Ministry of Environment at 1 km2 spatial categories for biodiversity conservation existing in resolution. Restricting our analysis to complete re - Andalusia by September of 2017: (1) national PAs gional censuses between the years 2010 and 2014 (N = 192), including the 8 regional designation cate- resulted in 617 species. Of these, the 126 species that gories: Natural Monument, Protected Landscape, were included in the Andalusian Register of Threat- Nature Site, National Park, Nature Park, Peri-urban ened Species under the categories Extinct, Endan- Park, Nature Reserve, and Private Nature Reserve gered, or Vulnerable (regionally threatened species (Table S1 in the Supplement at www. int-res. com/ or RTSs; Andalusian Government 2017b), and the 33 articles/suppl/ n035p125 _ supp.pdf ); (2) SCIs (N = 27); species that were included in the IUCN Red List (3) SACs (N = 163); (4) SPAs (N = 63); (5) biosphere under the categories of Regionally Extinct, Critically reserves (BRs; N = 9); (6) Ramsar sites (i.e. wetlands 128 Endang Species Res 35: 125–139, 2018

of international importance as defined by the Ramsar The protected polygon layer was then intersected Convention) (N = 25); and (7) World Heritage Sites with the 2 threatened species’ richness layers after (WHSs; N = 1). The regional terrestrial PA network dissolving them by the number of different threat- can be considered almost complete as analysed here, ened species present in each 1 km2 plot and counting notwithstanding possible future additions or mod- them. A Spearman rank correlation test was then ifications due to, for instance, global change (Ruiz- performed to ascertain the degree of relationship Mallén et al. 2015) or managerial regime (EEC 1992). between RTS and GTS richness and number of PA overlapping designation categories after checking

the non-normality of the original and log10-trans- Data analysis formed variables using SPSS software for an α = 0.05. A methodological summary flux diagram is shown PA layers were clipped against the Andalusian in Fig. 2. land territory layer (IGN 2011), to select just the ter- Management effort was estimated in a generic restrial PAs present in the region. The 7 PA category manner by asking the regional government’s man- layers were unioned and the resulting protected agers whether entire PA designation categories were polygons ≥100 ha were selected (n = 404) in order to generally actively managed and allocating their re - minimise layer intersection errors (mostly smaller sponses (actively managed vs. not actively managed) than that area) and also in order to obtain protected to each analysed PA designation category in the polygons of at least equal size to the species’ resolu- region. All GIS calculations were done using ArcGIS tion data. The layers were dissolved later on to calcu- v.10.3 (ESRI 2014) in the ETRS89 UTM 30N late the total PA in the region. The complete PA layer projection. and each of the 7 PA designation category layers were then intersected with the dissolved RTSs and GTSs layers showing their regional distribution, as RESULTS well as with the HCI layer. PA area and percentage calculations by designation category do not account Official protection of threatened species for overlaps among PA designation categories, so they should be read individually for each category. One hundred and twenty-six species in Andalusia Broad ecosystem types were estimated from land were regionally threatened and/or globally threat- use-land cover data. To determine the main ecosys- ened. Of these, 119 species were RTSs and 33 were tem type at each level of analysis, the Corine Land GTSs (Table S2 in the Supplement). Seven GTSs Cover (CLC) 2012 vector layer (Copernicus Land were not included in the Andalusian Register of Monitoring Services 2016) was clipped against the Threatened Species: Aythya ferina, Eryngium galio- Andalusian territorial layer (IGN 2011), and the com- ides, Galium viridiflorum, Genista ancistrocarpa, plete PA layer, RTS and GTS layers, and descriptive Prunus ramburii, Santolina elegans, and Succisella statistics were computed. In order to rank every CLC andreae-molinae. There was no statistically signifi- subclass according to its importance to threatened cant correlation between the regional and global cat- species, a unit-less ecosystem importance index (EII) egories of threat to the species in both subsets (RTS was produced for each level-three CLC subclass: EII and GTS). The regional area of occupancy of GTSs ∑ = (Nn × An), where N is the number of threatened was negatively correlated with their global degree of species and A is the percentage of the area occupied threat (rS(31) = −0.512; p = 0.02). by that number of species in CLC subclass n. The degree of coincidence in threat categories between common species in both subsets (RTSs and Representation of threatened species in the GTSs) and between their degree of threat and their regional PA network areas of occupancy were assessed via Spearman rank correlation tests, for a significance level of 0.05, PAs covered approximately 58% of the area of after normality checks of the original and log10-trans- occupancy of RTSs and 81% of the area of occupancy formed variables. Legal protection afforded to the of GTSs (Fig. 3). Species’ richness in 1 km2 plots var- 404 protected polygons of the complete PA layer was ied from 1 to 12 species for RTSs and from 1 to 4 spe- calculated by counting the number of overlapping cies for GTSs. PA coverage was high (>80%) for all PA designation categories in each polygon, as done plots containing ≥3 RTSs (Table S3 in the Supple- previously (Rodríguez-Rodríguez et al. 2015b, 2016). ment). PA coverage in GTSs’ plots was also high Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 129

Fig. 2. Methodological flux diagram for the present study. AHITB: Area of High Importance for Threatened Biodiversity, AHITS: Area of High Importance for Threatened Species, BR: Biosphere Reserve, EII: Ecosystem Importance Index, GTS: globally threatened species, HCI: Habitat of Community Interest, PA: Protected Area, RTS: regionally threatened species, SAC: Special Area of Conservation, SCI: Site of Community Importance, SPA: Special Protection Area, WHS: World Heritage Site

Area of occupancy Area of occupancy in (>79%) and increased with GTSs richness. All the outside regional PAs (ha) regional PAs (ha) plots having 3 or 4 GTS were included in the regional PA network (Table S4 in the Supplement). 100 The areas of occupancy of RTSs and GTSs cover 14202 1.8 times and 2.5 times more relative area inside 80 76502 regional PAs than in the rest of the territory, respectively. The total and detailed distribution of the area 60 occupied by threatened species by PA designation category in the region is shown in Table 1. %

40 61507

103536 Representation of threatened species in broad 20 ecosystem types

Diverse ‘agricultural’ ecosystems (CLC Class 2) 0 covered most of the region (Table S5 in the Supple- Regionally threatened Globally threatened species species ment). In turn, ‘forest and semi-natural’ ecosystems Fig. 3. Area of occupancy of regionally threatened species (CLC Class 3) extended over the largest part of the and globally threatened species inside and outside Andalu- region’s PA network (Table S6 in the Supplement). sian protected areas (PAs) The majority of the area of occupancy of RTSs and 130 Endang Species Res 35: 125–139, 2018

GTSs lay in ‘agricultural areas’, chiefly on non-irrigated arable land (25 and 22%, respectively; Tables S7 & S8 in the Supple- ment). Table 2 shows the representation of major ecosystems across these assessment units. The distribution of land uses–land covers according to threatened species’ richness is shown in Tables S9 & S10 in the Sup- plement. The most important ecosystems for RTSs and GTSs according to the EEI coincided: ‘wetlands’, chiefly ‘salines’

) (EIIRTS = 2904; EIIGTS = 466) and ‘inland marshes’ (EIIRTS = 2224;

81.24 2.18 EIIGTS = 296; Tables S11 & S12 in the Supplement). They covered most of the area of the species-richest plots (up to 100% for the richest plots). Of the 5 most important ecosystems for threatened species according to the EII, 2 were ‘wetlands’ and 3

(GTSs) in Andalusia. SAC: Special Area of were ‘agricultural areas’. In contrast, ‘forest and semi-natural’

28 174 37.21 20.22 ecosystems had a medium to low importance as threatened spe- 44 009 58.13 2.66 47 884 63.25 2.09 14 900 19.68 21.41 cies-rich areas. 44 949 59.37 2.95

Representation of HCIs in the regional PA network occupancy occupancy of occupancy including

Over 52% of the extent of the 71 HCIs in Andalusia was in - area Area of Area of area Total Protected area

(%) (ha) in region (%) in PAs protected (%) GTSs (%) cluded in the regional PA network. This amounts to 62% of the PA

5.01 0.58 50 412 66.59 3.10 network area (Table 3). Proportionally, the most important PA

4.21 0.04 3363 4.44 1.37 designation category for HCIs was SCIs, with 75% of the SCI area incorporating HCIs. WHSs and SCIs included proportionally 2.84 times and 2.32 times more HCI area than the regional HCI coverage, respectively.

Protection afforded to threatened biodiversity by the regional PA network

Most regional protected area was protected by 4 designation categories, although nearly the same proportion was protected by just 1 legal category (Fig. 4). The legal protection of regional HCIs followed a similar pattern (Fig. 5). The number of GTSs was not correlated with the number of overlapping protection categories, nor was the number of RTSs, although this correlation was nearly

2.06 103 536 57.51 3.67 0.86 61 507 significant (rS(50) = 0.259; p = 0.06). b Conservation, Special Protection SCI: Site of Community Importance, Area SPA: Fig. 6 shows the spatial overlap between threatened species’ richness and the number of overlapping PA designation categories. Considering that, in general (although not always), national

Accounting for overlaps PAs, SACs, BRs, Ramsar sites and WHSs all had active manage- b 69 587 0.79 1.08 17 813 9.89 25.60 0.90 139 325 1.59 0.42 36 731 20.40 26.36 0.32 ment in place, 75.2% of the regional PA can be considered 1 655 900 18.90 0.83 73 144 40.63 4.42 0.50 2 285 920 26.09 0.87 76 162 42.30 3.33 0.55 a

1524780 17.41 0.81 70 826 39.34 4.64 0.51 actively managed. Thus, 74.9 and 77.0% of areas occupied by a RTSs and GTSs in regional PAs (not considering overlaps) were in managed PAs (Table 1). SCIs and SPAs do not normally have a a active management, although some of them do, especially when they overlap with any of the ‘managed’ designation categories.

a Not accounting for overlaps among PA designation categories, 74.6% of the HCI area in regional PAs can be considered actively Managed PA categories; Managed PA Area Coverage Area of area Area of Total Protected (%) in region (%) (ha) protected in PAs (%) RTSs (ha) in region occupancy occupancy of occupancy including SCIs 244 843 2.79 0.12 10 297 5.72 SPAs 1 628 680 18.59 0.93 81 568 45.31 SACs Ramsar sites Heritage Sites World Biosphere reserves Andalusia 2 817 830 32.17 Distribution (ha) (n = 404) PAs (n = 126) RTSs a GTSs (n = 33 National PAs

Table 1. Descriptive statistics on protected regionallyTable areas threatened (PAs), and globally threatened species (RTSs) species managed (Table 3). Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 131

Table 2. Representation of broad ecosystem types across assessment units, AHITBs: new candidate PAs in percentage. PA: protected area, RTS area: area occupied by regionally threatened species, GTS area: area occupied by globally threatened species HCIs only spatially coincided with RTSs or GTSs on 1 site of 111 ha (of a Assess- Broad ecosystem type 200 ha AHITS) to the north of Cordoba ment Agricultural Forest Wetland Other Artificial unit eco- eco- eco- freshwater eco- province. There is a total of 111 systems systems systems ecosystems systems AHITSs, ranging from 100 to 400 ha in size (Fig. 7). The biggest one, of 400 ha, Region 56.99 38.66 0.76 0.93 2.66 was also to the north of Cor doba Regional PAs 29.29 66.48 2.13 1.64 0.45 province and contained Otis tarda (a RTS area 51.00 31.97 11.89 3.14 1.57 GTS area 44.27 32.15 19.25 3.05 1.28 GTS) and Rhino lo phus ferrum equinum (an RTS). It bordered with other unpro-

Table 3. Descriptive statistics on protected area (PA) coverage and extent of habitats of community interest (HCIs) in Andalusia. SAC: Special Area of Conservation, SCI: Site of Community Importance, SPA: Special Protection Area

Distribution PAs (n = 404) HCIs (n = 71) Area Coverage in Extent in Extent in PAs Extent in PAs Extent in PA (ha) region (%) region (%) (ha) (%) category (%)

Andalusia 2 817 830 32.17 37.68 1 733 274 61.51 61.51 National PAsa 1 655 900 18.90 13.12 1 149 489 40.79 69.42 SCIs 244 843 2.79 2.08 182 429 6.47 74.51 SPAs 1 628 680 18.59 12.48 1 093 147 38.79 67.12 SACsa 2 285 920 26.09 16.92 1 482 095 52.60 64.84 Ramsar sitesa 139 325 1.59 0.83 72 632 2.58 52.13 Biosphere reservesa 1 524 780 17.41 11.39 998 172 35.42 65.46 World Heritage Sitesa 69 587 0.79 2.29 37 342 2.26 53.66

aManaged PA categories

Protected area 47.50 HCI area 50 50

39.93 38.3 40 40

30 30 24.81 % % 20 20 15.66

11.27

10 6.41 10 5.94 4.18 1.63 2.46 0.79 1.13

0 0 162345 0612345

Cumulative protected area designations Cumulative protected area designations

Fig. 4. Percentage of Andalusian protected area covered by Fig. 5. Percentage of habitat of community interest (HCI) overlapping legal designations area covered by overlapping protected area categories in Andalusia 132 Endang Species Res 35: 125–139, 2018

than twice the global PA coverage, at 14.7% (Bhola et al. 2016) and nearly doubling international PA coverage targets for terrestrial ecosystems (17%) by 2020 (CBD 2010). The network is positively skewed to wards protection of threatened species, as ex pected, and includes the areas richest in threatened species in the region for both subsets. This is encouraging, as PAs have been found to reduce the extinction risk of threatened species (Butchart et al. 2012). However, substantial improvements can be made with respect to the spatial representation of threatened biodiversity in the regional network of PAs, especially for RTSs and HCIs. Even though the areas occupied by RTSs and GTSs cover a very small fraction of the regional area, a substantial proportion of this area is currently afforded no specific legal protection. Never- theless, the disjoint and geographi- cally scattered location of most of that unprotected area makes it challenging to endow it some legal and, especially, managerial protection regime. Our results are in accordance with a number of studies that show insufficient representation of threatened species in PAs globally (Gaston et al. 2008, Bhola et al. 2016), across Spain (Muñoz-Rodríguez et al. 2016) and in southeastern Andalusia (Mendoza- Fernández et al. 2014). Fig. 6. Spatial overlap between cumulative legal protection of protected areas Ramsar sites occupy comparatively and (a) regionally threatened species’ richness and (b) globally threatened small areas in Andalusia, but are by species’ richness in Andalusia far the most representative PA category for RTSs and GTSs. This was un - tected plots of 1400 ha in total that also contained Cir- expected, given the relatively broad designation cus pigargus (an RTS). objectives of such areas (Ramsar Convention 1971) and that some other PA designation categories were specifically created for threatened biodiversity (e.g. DISCUSSION SCIs, SACs and SPAs). This result underpins the importance of wetland ecosystems for the conserva- Representation of threatened biodiversity in the tion of threatened species in the region and may be regional PA network explained by the unique nature, vital territorial func- tion, and restricted distribution of wetlands in An da - The Andalusian terrestrial PA network can be lusia (Andalusian Government 2002). However, our considered mature and well developed in quantita- re sults might have been biased by greater inclusion tive, general terms. It is now nearly 30 yr old and of wetland biodiversity in the original threatened covers over one-third of the regional territory, more species’ dataset. Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 133

a very good predictor of threatened species’ richness in the region, especially for GTSs. Incomplete ecological information prior to PA designation and multiple and changing PA designation criteria over time (Múgica et al. 2002) are likely to have influenced an improvable coverage of threatened species by many old PAs designated in the late 1980s in Andalusia, especially by ‘National’ PA designation categories (Andalusian Government 1989). It is worth noting that current in situ biodiversity conservation criteria and definitions largely derive from the 1992 Río Summit and the related Con- vention on Biological Diversity (CBD 1992) and subsequent amendments, notably the Strategic Plan for Biodi- Fig. 7. New candidate protected areas in the Andalusian region (≥100 ha) according to the number of threatened species (AHITS: Area of High Importance versity 2011−2020 (CBD 2010). for Threatened Species) and extent of habitats of community interest (AHITB: Assuming that legal designation of Area of High Importance for Threatened Biodiversity) PAs, whatever their protection strin- gency, is insufficient for effective conservation without appropriate man- Almost half of the regional HCI area is outside the agement is widely assumed (Hockings et al. 2006, regional PA network. However, the large area cov- Dudley 2008, Bhola et al. 2016). However, manage- ered by regional HCIs makes it challenging to com- ment of the Andalusian PA network was assessed pletely protect them (Cabello et al. 2009). Global habi- here in very general terms and remains a pending tat-scale data were not available, but representation research task. of terrestrial biodiversity in PAs at broader scales, such as ecoregions, could be much improved since only 10% of ecoregions enjoy legal protection for Representation of threatened species by broad ≥50% of their area (Bhola et al. 2016). The fact that ecosystem types HCIs are best represented in WHSs was also unex- pected, given the broad designation objectives of The regional PA network is positively skewed to- such a PA category (UNESCO 1972) and the HCI- spe- wards the protection of wetlands, water bodies and cific designation of SCIs and SACs (EEC 1992). Actu- forest areas, and negatively skewed towards the pro- ally, SCIs were less important for HCI protection than tection of agricultural areas. The PA network in - WHSs. However, this can be explained by the fact that cludes proportionally nearly 3 times more wetland WHS data comes from just one exceptionally biologi- area than wetland representation in the regional ter- cally rich site (Doñana wetland), which is also a SAC. ritory. Wetlands are even more disproportionally represented in the areas occupied by RTSs (almost 13 times more wetland area than in the whole of Anda- Protection of threatened biodiversity afforded by lusia) and GTSs (more than 25 times more wetland the regional PA network area). Wetlands are threatened ecosystems globally (Turner et al. 2000) as well as in the Mediterranean The regional PA network can be considered as basin (Beltrame et al. 2012). It was estimated that highly legally protected and afforded high theoreti- Spain has lost around 60−70% of its historical wet- cal management effort. The networks’ legal protec- land area to agricultural, urban and public health tion seems to be in accordance with RTS richness, development, with similar figures for Andalusia which suggests coherent legal effort to protect RTSs. (Andalusian Government 2002). Today, Andalusia Nevertheless, the degree of legal protection assessed has the highest number and area of protected wet- as the number of overlapping PA categories was not lands of all the Spanish regions (Nieto 2012). The ter- 134 Endang Species Res 35: 125–139, 2018

ritorial value of Andalusian wetlands for threatened diversity in Mediterranean contexts, and suggest the biodiversity is likely growing, as climate predictions importance of maintaining environmentally friendly forecast increased aridity in the northern Mediter- land uses and practices aimed at threatened species ranean basin and, more markedly, in the central and in these ecosystems, both inside and outside PAs. southern Iberian Peninsula (Barredo et al. 2016). Acknowledging the importance of wetlands for regional biodiversity, a Wetland Plan was produced Opportunities for enhanced biodiversity protection with the purpose of conserving wetlands’ ecological integrity and promoting their sustainable use (Anda - The number of threatened species outside PAs is lusian Government 2002). The 25 Ramsar sites in generally low to medium in relative terms. Neverthe- Andalusia have also been designated as SPAs (20 of less, high absolute numbers of RTSs occur outside them) and/or national PAs (all sites; Andalusian Gov- the regional PA network. This is worrisome, as the ernment 2016c), chiefly Nature Reserves and Nature Mediterranean biogeographical region, which in - Sites that, together with National Parks, are the most cludes the whole of Andalusia, is territorially more legally restrictive regional designation categories. dynamic than the Atlantic region in terms of land Thus, these vital and spatially restricted ecosystems use– land cover changes in Spain (Martínez-Fernán- for threatened species are theoretically highly pro- dez et al. 2015). Actually, land cover changes towards tected in Andalusia. land abandonment, urbanisation, increased land-use Forests are of limited importance regarding the intensity and over-exploitation of water resources spatial occurrence and richness of threatened spe- have been common in the recent history of Euro- cies. In contrast to global figures, where PAs contain- Mediterranean regions, and very widespread in ing natural or recovering land uses were more biodi- Spain between the late 1980s and mid-2000s (Stell - verse than those having human-dominated land uses mes et al. 2013). Even though massive residential (IUCN 2016), 5 of the 6 broad ecosystem types richest and infrastructure construction declined abruptly in in threatened species (from 8 to 12 species, for RTSs, Spain from 2008 as a consequence of the burst of the and from 3 to 4 species, for GTSs) can be considered housing bubble, economic recovery is posing a risk of human-modified ecosystems in Andalusia, including natural habitat transformation, especially in the most agricultural areas. This may be explained by the dynamic areas around the main cities and coastal global character of the IUCN study and the excep- zones (Jiménez 2010, Alfonso et al. 2016). Therefore, tional species richness of tropical and equatorial legal protection of the identified AHITSs and AHITB forests, both in absolute terms and in relative terms should be considered, not only to protect threatened related to threatened species (Myers et al. 2000). In biodiversity but also to increase landscape perme- the historically modified Mediterranean ecosystems, ability in a context of foreseen shifts in species’ distri- environmentally heterogeneous cultural landscapes bution ranges due to global change (Araújo et al. have been found to have high levels of biological 2011). In this regard, some regional initiatives such diversity (Schmitz et al. 2005) and contain semi- as the draft master plan for the improvement of natural habitats of high conservation value (Araújo et ecological connectivity in Andalusia (Anda lusian al. 2007): Mediterranean ecosystems ranked second Government 2016d) aim to anticipate and revert the in threatenend species’ richness after tropical forests irreversible consequences of those changes by sus - (Myers et al. 2000). Human interventions in ecosys- tain ably managing the whole regional territory. tems seem to follow opposite trends but have similar The mismatch between national (or regional) threat consequences in tropical areas and in Mediterranean category criteria and international threat criteria areas. Whereas in tropical areas, conversion of natu- (IUCN 2012a,b) makes it possible that some taxa of lit- ral land covers to human-dominated ecosystems is a tle global concern might be regionally threatened and major cause of biodiversity loss, in Euro-Mediterran- that conservation actions are applied to them. It also ean areas, abandonment of traditional extensive uses makes the opposite and most worrisome possible: that of the land, such as non-irrigated farming or free- some GTSs are not considered RTSs because they roaming livestock, is leading to natural succession might be regionally abundant and are thus not af- towards forest habitats and subsequent loss of forded specific conservation actions by the regional sparsely vegetated habitats and homo genisation of government despite their global degree of threat. For associated biodiversity (Martínez- Fernández et al. instance, of the 7 such species identified here, 4 are 2015, Herrando et al. 2016). Our results underpin the globally Endangered species and 3 are Vulnerable value of human-made ecosystems for threatened bio- species according to internationally agreed criteria Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 135

(IUCN 2017). Therefore, they should be given protec- Other methodological issues such as the small scale tion priority and be included in the Andalusian Regis- of CLC data or inherent geometric and thematic ter of Threatened Species. The application of threat errors when using spatial data techniques may have categories based on administrative boundaries is dis- influenced our results to some degree (Martínez-Fer- couraged by international guidelines (Gärdenfors et nández et al. 2015). Moreover, the kind of data pro- al. 2001), as threat level is very scale-dependent vided for species by the regional administration, (IUCN 2012a,b). Thus, the current approach to taxa where census points and areas of occupancy are conservation followed by the national and, especially, depicted at 1 km2 resolution, makes it most likely that the regional governments in Spain could probably be the actual areas of occupancy are smaller than shown made more economically efficient and ecologically here for many species of limited mobility, and thus meaningful by considering broader spatial scales. the regional distribution of some threatened species The easternmost part of the Doñana PA, in the may be overestimated. Finally, completeness, consis- southwestern part of the region, stands out in terms tency and accurateness of the official census data of the density of threatened species, with absolute provided by the Regional Ministry of Environment maximal values of 12 RTSs and 4 GTSs by 1 km2 cannot be assured, given the extent of the Andalu- plots. Previous studies highlighted high densities of sian territory, and the large number of species, habi- threatened species in the Doñana PA, and also in tats, monitoring protocols and people involved, and Sierra Nevada National Park and northern Andalu- should thus be assumed to be the best available, most sia, respectively (Múgica et al. 2010). This high con- comprehensive existing data for the region. Data for centration of threatened species at the periphery of a few emblematic species such as the grey wolf or the the PA would suggest some eastward expansion of Iberian lynx were not included in the official data- the Doñana PA (under various possible designation base, possibly due to their high sensitivity, although categories) that could act as a buffer against external both species are closely and regularly monitored in pressures to the PA. Such pressures are currently the region. However, the large effort made by the serious and manifold (Carmona 2012, Carmona et al. regio nal government to monitor such high numbers 2012, Carmona & Fuentelsaz 2013, 2015, WWF 2017). of threatened species and habitats at the level of Thus, extreme caution should be exercised before detail shown here, as well as to compile and provide authorising any potentially impacting activity in or re lated data is uncommon, not only in Spain, where near this exceptionally valuable area for threatened monitoring of protected biodiversity is very variable species and, whenever possible, other less environ- among regions and generally deficient (Rodríguez- mentally valuable and vulnerable locations for such Rodríguez et al. 2015a), but also European-wide activities should be sought. (Davis et al. 2014) and globally (Gaston et al. 2008, IUCN 2012a), and merits recognition.

Methodological limitations CONCLUSIONS Methodological improvements to the study’s accuracy could include: (1) analysing species’ protection The Andalusian PA network has broad territorial according to PA’s legal zones, (2) assessing protec- coverage and is highly legally and managerially pro- tion of particular species or habitats according to tected in general, quantitative terms. It also includes each PA’s designation objectives (e.g. bird species in the regional areas that are richest in threatened spe- SPAs), or (3) determing population trends for species cies. However, our analysis suggests that representa- inside and outside PAs according to similar bio-phys- tion of RTSs and HCIs can be improved and that legal ical covariates. protection can be better allocated according to It should be noted that PAs <100 ha were discarded threatened species’ richness, chiefly of GTSs. Non- from this analysis, thus slightly underestimating rep- irrigated arable land was the most important ecosys- resentation and protection of threatened biodiversity, tem type for threatened species, in terms of areas of and making it possible that some small parts of the occupancy. Wetlands and some agricultural areas AHITSs or AHITB are actually protected but not stood out as the broad ecosystem types richest in reflected in our analysis. Also, managerial protection threatened species in Andalusia. Therefore, their as assessed here is just a rough generalisation of proper delimitation, protection (where needed) and actual management activities carried out in regional management are of paramount importance and PAs and is, in contrast, most likely overestimated. should be granted. 136 Endang Species Res 35: 125–139, 2018

Ramsar sites were especially important for the con- whereby the Andalusian Protected Area Inventory is cre- servation of RTSs and GTSs, but less important for ated, and additional protection measures are established). www.juntadeandalucia.es/boja/1989/60/1/ content/24/10/ the conservation of threatened habitats. One-hun- 16 (accessed 24 Jul 2017) dred and eleven AHITSs and one AHITB were iden- Andalusian Government (2002) Plan Andaluz de humedales tified. The designation as PAs of at least the biggest (Andalusian wetland plan). Consejería de Medio Ambi- AHITB and AHITS in the northern part of the region ente, Junta de Andalucía, Sevilla (in Cordoba province) as well as the eastward expan- Andalusian Government (2014) Medio ambiente en Andalucía. Datos básicos 2014 (The environment in Andalusia. Basic sion of Doñana PA also emerge as priorities from this data 2014). www.juntadeandalucia.es/medio ambiente/ analysis. Finally, the inclusion of the 7 GTSs not cur- portal_web/rediam/productos/Publica ciones/ datos_ rently included in the Andalusian Register of Threat- basicos_2014/datosbasicos2014html/index. html (ac cessed ened Species is paramount. 9 Feb 2017) This study is a snapshot of the status of the regional Andalusian Government (2015) Hábitats de interés comunitario terrestres en Andalucía. Metodología para gen- PA network up to 2017. It is thus advisable to take eración de cartografía HICs 2015 (Terrestrial habitats of ap propriate measures to improve the network ac - community interest in Andalusia. Methodology for the cording to today’s data and environmental context. production of HCI maps 2015). www.junta de anda lucia. Global change effects will likely modify the distribu- es/medioambiente/portal_web/rediam/documen tos/ tion and density of threatened biodiversity. Scien- vegetacion/Metodologia_HIC2015.pdf (accessed 9 Feb 2017) tists, managers and decision-makers should be ready Andalusian Government (2016a) La RENPA − Red de Espa- to react soundly and promptly to such changes by cios Naturales Protegidos de Andalucía (The Andalusian adapting the boundaries of PAs, and their legal spec- Protected Area Network). www.juntadeandalucia. es/ ifications and managerial practices to promote effec- medioambiente/site/portalweb/menuitem.7e1cf46ddf59 bb227a9ebe205510e1ca/?vgnextoid=0cbb7abc83414010 tive conservation. Natural and semi-natural areas VgnVCM1000000624e50aRCRD&vgnextchannel=3bdd6 surrounding existing PAs and ecological corridors 1ea5c0f4310VgnVCM1000001325e50aRCRD (accessed between them will become essential to facilitate adap- 24 Jul 2017) tation and survival of threatened biodiversity in the Andalusian Government (2016b) Hábitats de interés comu- face of global change (Araújo et al. 2011). nitario de Andalucía. Publicación 2015 (Habitats of community interest). www.juntadeandalucia.es/medioambi ente/site/rediam/menuitem.04dc44281e5d53cf8ca78ca7 Acknowledgements. This paper recognises contributions 31525ea0/?vgnextoid=a917d2aa40504210VgnVCM1000 through the ‘sequence-determines-credit’ approach. We ac - 001325e50aRCRD&vgnextchannel=7b3ba7215670f210V knowledge and thank Pepe Perea, Francisco Marín, Yo lan da gnVCM1000001325e50aRCRD&lr=lang_es (accessed 21 Gil and José Manuel Moreira, from the Andalusian Environ- Jul 2017) mental Information Network (REDIAM), Andalusian Min- Andalusian Government (2016c) Humedales de Andalucía en istry of Environment and Territorial Planning, for providing la lista Ramsar (Andalusian wetlands on the Ramsar list). the official data on threatened species in Andalusia, and also www.juntadeandalucia.es/medioambiente/site/ portal for providing advice on how to treat and interpret species web/menuitem.7e1cf46ddf59bb227a9ebe205510 e1ca/ and habitat data. We also thank Gonzálo González, from the ?vgnextoid=4efabfe08a7a5010VgnVCM1000000624e50 a Andalusian Ministry of Environment, for providing useful RCRD &vgnextchannel=a22ca0d0851f4310Vgn VCM 2000 insights on a draft version of the manuscript, and Milagros 000624e50aRCRD (accessed 24 Jul 2017) Pérez, for clarifying aspects of PA management in the region. Andalusian Government (2016d) Plan director para la me - Finally, we thank 3 anonymous reviewers whose remarks jora de la conectividad ecológica en Andalucía. Informe helped us to substantially enhance the content of this study. de síntesis. Septiembre 2016 (Draft master plan for the D.R.R. was funded for this study by the Spanish Ministry of improvement of ecological connectivity in Anda lusia). Economy and Competitiveness in the framework of the www.juntadeandalucia.es/medioambiente/ portal_ SOSTPARK project (CSO2014-54611-JIN). web/web/temas_ambientales/biodiversidad/6_habitats _y_ecosistemas/0_plan_conectividad/cea_FINAL_doc0_s intesis_septiembre_2016.pdf (accessed 5 Oct 2017) LITERATURE CITED Andalusian Government (2017a) Seguimiento de la biodiver - sidad (Biodiversity monitoring). www.juntadeandalucia. Alfonso C, Avellaner J, Estévez R, García A and others es/medioambiente/site/portalweb/menuitem.220de8226 (2016) Sostenibilidad en España 2016. Cumplimiento de 575045b25f09a105510e1ca/?vgnextoid=c9984df2889273 los objetivos de desarrollo sostenible de Naciones 10VgnVCM2000000624e50aRCRD&vgnextchannel=efa9 Unidas. SOS16 (Sustainability in Spain 2016. Meeting of 6c3b0ef95310VgnVCM1000001325e50aRCRD (acces sed United Nations’ sustainable development goals. SOS16). 5 Oct 2017) Asociación Observatorio de la Sostenibilidad Andalusian Government (2017b) Listado y catálogo de flora y Andalusian Government (1989) LEY 2/1989, de 18 de julio, fauna silvestre amenazada, con presencia regular, en paso por la que se aprueba el Inventario de Espacios Naturales u ocasional en Andalucía (List and register of threatened Protegidos de Andalucía, y se establecen medidas adi- wild flora and fauna with regular, occasional or migratory cionales para su protección (Law 2/1989, from 18th of July, presence in Andalusia). www.juntade andalucia. es/medio Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 137

ambiente/site/portalweb/menuitem. 7e1cf46ddf59bb 227 Carmona J, Fuentelsaz F (2015) En el XVII aniversario de la a9ebe205510e1ca/?vgnextoid=a23f3e 9f6127 c410 Vgn catástrofe de Aznalcóllar, WWF reclama a la Junta que VCM1000001325e50aRCRD&vgnext channel=91aac c879 no reabra la mina (In the XVIIth anniversary of the a47c410 VgnVCM1000001325e50aRCRD (accessed 3 Oct Aznalcollar catastrophy, WWF demands the Junta not to 2017) reopen the mine). WWF Info Doñana 184: 1−3 Andalusian Government (2017c) Descargas de información Carmona J, Fuentelsaz F, Hernández E (2012) El robo del ambiental. Patrimonio natural. Espacios protegidos. agua en Doñana. Cinco casos de estudio. WWF informe RENPA (Environmental data download. Natural heritage. 2012 (The stealing of water in Doñana. Five case studies. Protected areas. RENPA). http:// descargasrediam. cica. WWF report 2012). WWF/ADENA es/repo/s/RUR?path=%2F07_PATRIMONIO _ NATURAL CBD (Convention on Biological Diversity) (1992) Text of the % 2F01_ESPACIOS_PROTEGIDOS%2F01_RENPA %2F CBD. www.cbd.int/convention/text/ (accessed 24 Jul 2017) 00_RENPA%2FRENPA_actual%2FInf Geografica%2FInf CBD (Convention on Biological Diversity) (2010) Strategic Vectorial%2FShapes (accessed 27 Sep 2017) plan 2011−2020: Aichi biodiversity targets. www.cbd. int/ Araújo MB, Lobo JM, Moreno JC (2007) The effectiveness of sp/targets/ (accessed 24 Jul 2017) Iberian protected areas in conserving terrestrial biodi- Copernicus Land Monitoring Services (2016) Corine Land versity. Conserv Biol 21: 1423−1432 Cover: CLC 2012. http://land.copernicus.eu/pan-euro Araújo MB, Alagador D, Cabeza M, Nogués-Bravo D, pean/corine-land-cover/clc-2012/view (accessed 24 Jul Thuiller W (2011) Climate change threatens European 2017) conservation areas. Ecol Lett 14:484−492 Craigie ID, Baillie JEM, Balmford A, Carbone C, Collen B, Barredo I, Caudullo G, Dosio A (2016) Mediterranean habi- Green RE, Hutton JM (2010) Large mammal population tat loss under future climate conditions: assessing im - declines in Africa’s protected areas. Biol Conserv 143: pacts on the Natura 2000 protected area network. Appl 2221−2228 Geogr 75: 83−92 Davis M, Naumann S, McFarland K, Graf A, Evans D (2014) Beltrame C, Chazée L, Galewsky T, Perennou C (2012) Literature review: the ecological effectiveness of the Mediterranean wetlands: outlook. Mediterranean Wet- Natura 2000 network. Tech Pap No. 5/2014. European lands Observatory’s first technical report. Mediterranean Topic Centre on Biological Diversity, European Environ- Wetlands Observatory ment Agency, Paris Bhola N, Juffe-Bignoli D, Burguess N, Sandwith T, Kingston Dudley N (ed) (2008) Guidelines for applying protected area N (eds) (2016) Protected planet report 2016. How pro- management categories. IUCN, Gland tected areas contribute to achieving global targets for EC (European Commission) (2016) Natura 2000 barometer. biodiversity. UNEP-WCMC and IUCN, Cambridge and http://ec.europa.eu/environment/nature/ natura2000/ Gland barometer/index_en.htm (accessed 24 Jul 2017) Bland LM, Keith DA, Miller RM, Murray NJ, Rodríguez JP EEA (European Environment Agency) (2012) Biogeographical (eds) (2017) Guidelines for the application of IUCN Red regions in Europe. www.eea.europa.eu/data-and-maps/ List of ecosystems categories and criteria, version 1.1. figures/ biogeographical-regions-in-europe-1 (accessed IUCN, Gland 24 Jul 2017) Butchart SHM, Walpole M, Collen B, van Strien A and oth- EEC (European Economic Community) (1979) Council ers (2010) Global biodiversity: indicators of recent de - Directive 79/409/EEC of 2 April 1979 on the conservation clines. Science 328: 1164−1168 of wild birds. http: //eur-lex.europa.eu/legal-content/ES/ Butchart SHM, Schalermann JPW, Evans MI, Quader S and TXT/?uri=URISERV%3Al28046 (accessed 24 Jul 2017) others (2012) Protecting important sites for biodiversity EEC (European Economic Community) (1992) Council Direc- contributes to meeting global conservation targets. PLOS tive 92/43/EEC of 21 May 1992 on the conservation of nat- ONE 7: e32529 ural habitats and of wild fauna and flora. http: //eur-lex. Cabello J, Escudero A, Rodà F, Olano JM and others (2009) europa.eu/legal-content/en/TXT/?uri=CELEX %3A 31992 Matorrales esclerófilos, halonitrófilos y estepas continen- L0043 (accessed 24 Jul 2017) tales halófilas y gipsófilas. En: VV.AA. Bases ecológicas ESRI (Environmental Systems Resource Institute) (2014) preliminares para la conservación de los tipos de hábitat ArcInfo, version 10.3. ESRI, Redlands, CA de interés comunitario en España (Sclerophyllous scrub, Estévez R, Prieto F, Alfonso C (2016) Cambios de ocupación halonitrophyllous scrub and continental hallophyllous del suelo en la costa. Informe basado en datos del Corine and gypsophyllous steppes. In: Preliminary ecological Land Cover. COSTA16 (Land cover changes on the coast. foundations for conservation of habitats of community Corine Land Cover-based report. COSTA16). Obser va to - interest in Spain). Ministerio de Medio Ambiente, y rio de la Sostenibilidad. http://www.observatoriososteni- Medio Rural y Marino, Madrid. www.mapama.gob.es/es/ bilidad.com/documentos/COSTA%2016v01_a.pdf biodiversidad/temas/espacios-protegidos/09047122 80 20 f (accessed 24 Feb 2017) 271_tcm7-24196.pdf (accessed 16 Oct 2017) Gärdenfors U, Hilton-Taylor C, Mace GM, Rodríguez JP Carmona J (2012) 10 retos para salvar el Guadalquivir: prop- (2001) The application of IUCN Red List criteria at uestas de WWF. WWF informe 2012 (10 challenges to regional levels. Conserv Biol 15: 1206−1212 save the Guadalquivir river: proposals by WWF. WWF Gaston KJ, Jackson SF, Cantú-Salazar L, Cruz-Piñón G report 2012). WWF España (2008) The ecological performance of protected areas. Carmona J, Fuentelsaz F (2013) WWF pide que no se per- Annu Rev Ecol Evol Syst 39: 93−113 mita la inyección de gas en Doñana y se opone a nuevas Herrando S, Brotons L, Anton M, Páramo F and others (2016) campañas de investigación (WWF opposes gas injection Assessing impacts of land abandonment on Mediterran- in Doñana as well as new gas exploration campaigns). ean biodiversity using indicators based on bird and but- WWF Info Doñana 158: 1−5 terfly monitoring data. Environ Conserv 43: 69−78 138 Endang Species Res 35: 125–139, 2018

Hockings M, Stolton S, Dudley N (2006) Evaluating effec- Myers N, Mittermeier RA, Mittermeier CG, da Fonseca tiveness: a framework for assessing the management of GAB, Kent J (2000) Biodiversity hotspots for conservation protected areas, 2nd edn. IUCN, Gland priorities. Nature 403: 853−858 Hoekstra JM, Boucher TM, Ricketts TH, Roberts C (2005) Nieto I (2012) Plan Andaluz de humedales. 10o Aniversario de Confronting a biome crisis: global disparities of habitat la aprobación (Andalusian wetland plan. 10th anniver- loss and protection. Ecol Lett 8: 23−29 sary). Consejería de Agricultura, Pesca y Medio Ambi- IGN (Instituto Geográfico Nacional) (2011) Base Cartográfica ente, Junta de Andalucía, Sevilla. www.juntadeandalucia. Nacional 1: 200.000. BCN 1:200. Version 4.0 (Spanish Car- es/medioambiente/portal_web/servicios_generales/ doc _ tography Base 1: 200,000. BCN 1:200. Version 4.0). http:// tecnicos/publicaciones_renpa/memoria_humedales_10_ centrodedescargas.cnig.es/CentroDescargas/buscador aniversario/10_aniversario_pah.pdf (accessed 21 Jul 2017) Catalogo.do?codFamilia=BC200# (accessed 4 Nov 2016) Ramsar Convention (1971) The text of the Ramsar Conven- IUCN (2012a) IUCN Red List categories and criteria: version tion and other key original documents. http://archive. 3.1, 2nd edn. IUCN, Gland ramsar. org/cda/en/ramsar-documents-texts/main/ramsar/ IUCN (2012b) Guidelines for application of IUCN Red List 1-31-38_4000_0 (accessed 24 Jul 2017) criteria at regional and national levels, version 4.0. Rodríguez-Rodríguez D, Martínez-Vega J, Tempesta M, IUCN, Gland Otero-Villanueva MM (2015a) Limited uptake of pro- IUCN (2016) A global standard for the identification of key tected areas evaluation systems among managers and biodiversity areas, version 1.0. IUCN, Gland decision-makers in Spain and the Mediterranean Sea. IUCN (2017) The IUCN Red List of Threatened Species, ver- Environ Conserv 42: 237−254 sion 17-2. www.iucnredlist.org (accessed 26 Sep 2017) Rodríguez-Rodríguez D, Sánchez-Espinosa A, Schroder C, Jiménez LM (Dir.) (2010) Sostenibilidad en España 2010 Abdul Malak D, Rodríguez J (2015b) Cumulative pres- (Sustainability in Spain 2010). Observatorio de la Soste- sures and low protection: a concerning blend for Medi- nibilidad en España terranean MPAs. Mar Pollut Bull 101: 288−295 Jiménez LM (Dir.) (2012) Sostenibilidad en España 2012. Rodríguez-Rodríguez D, Rodríguez J, Abdul Malak D (2016) Capítulo especial energía sostenible para todos. 2012 Development and testing of a new framework for rapidly Año Internacional de la Energía (Sustainability in Spain assessing legal and managerial protection afforded by 2012. Special chapter sustainable energy for all. 2012 marine protected areas: Mediterranean Sea case study. International Year of Energy). Ministerio de Agricultura, J Environ Manage 167: 29−37 Alimentación y Medio Ambiente, Madrid Ruiz-Mallén I, Corbera E, Calvo-Boyero D, Reyes-García D Juffe-Bignoli D, Burguess ND, Bingham H, Belle EMS and (2015) Participatory scenarios to explore local adaptation others (2014) Protected planet report 2014: tracking to global change in biosphere reserves: experiences from progress towards global targets for protected areas. Bolivia and Mexico. Environ Sci Pol 54: 398−408 UNEP-WCMC, Cambridge Schmitz MF, Pineda FD, Castro H, de Aranzabal I, Aguilera Laurance WF, Useche DC, Rendeiro J, Kalka M and others P (2005) Cultural landscape and socioeconomic struc- (2012) Averting biodiversity collapse in tropical forest ture. Environmental value and demand for tourism in a protected areas. Nature 489:290−294 Mediterranean territory. Consejería de Medio Ambiente, Martínez-Fernández J, Ruiz-Benito P, Zavala MA (2015) Junta de Andalucía, Sevilla Recent land cover changes in Spain across biogeograph- Spanish Government (2007) Ley 42/2007, de 13 de diciembre, ical regions and protection levels: implications for con- del patrimonio natural y de la biodiversidad (Law 42/2007, servation policies. Land Use Pol 44:62−75 from 13th of December, on natural heritage and biodiver- Médail F, Quézel P (1999) Biodiversity hotspots in the Medi- sity). www.boe.es/buscar/act.php?id=BOE-A-2007-21490 terranean Basin: setting global conservation priorities. (accessed 24 Jul 2017) Conserv Biol 13: 1510−1513 Spanish Government (2011) Real decreto 139/2011, de 4 de Mendoza-Fernández A, Pérez-García FJ, Martínez-Hernán- febrero, para el desarrollo del Listado de Especies Sil- dez F, Medina-Cazorla JM and others (2014) Threatened vestres en Régimen de Protección Especial y del Catál- plants of arid ecosystems in the Mediterranean Basin: a ogo Español de Especies Amenazadas (Royal decree case study of the south-eastern Iberian Peninsula. Oryx 139/2011, from 4th of February, for the development of 48: 548−554 the List of Wild Species under a Special Protection Re - Múgica M, de Lucio JV, Martínez C, Sastre P, Atauri- gime and of the Spanish Register of Threatened Species). Mezquida JA, Montes C (2002) Territorial integration of https:// www.boe.es/buscar/act.php?id=BOE-A-2011-3582 natural protected areas and ecological connectivity (accessed 5 Oct 2017) within Mediterranean landscapes. Consejería de Medio Spanish Government (2017) Resolución de 6 de marzo de Ambiente, Junta de Andalucía, Sevilla 2017, de la Dirección General de Calidad y Evaluación Múgica M, Martínez C, Gómez-Limón J, Puertas J, Atauri Ambiental y Medio Natural, por la que se publica el JA, de Lucio JV (2010) Anuario EUROPARC-España del Acuerdo del Consejo de Ministros de 24 de febrero de estado de los espacios naturales protegidos 2009. Fun- 2017, por el que se aprueban los criterios orientadores dación Fernando González Bernáldez, Madrid para la inclusión de taxones y poblaciones en el Catálogo Múgica M, Martínez C, Atauri JA, Gómez-Limón J, Puertas Español de Especies Amenazadas (Decision from 6th of J, García D (2014) EUROPARC-España. Anuario 2013 March of 2017, […] that passes the guidelines for inclu- del estado de las áreas protegidas en España. Fundación sion of taxa and populations in the Spanish Register of Fernando González Bernáldez, Madrid Threatened Species). www.boe.es/buscar/doc.php? id= Muñoz-Rodríguez P, Draper D, Moreno JC (2016) Global BOE-A-2017-2977 (accessed 5 Oct 2017) strategy for plant conservation: inadequate in situ conser- Stellmes M, Röder A, Udelhoven T, Hill J (2013) Mapping vation of threatened flora in Spain. Isr J Plant Sci SI:1−12 syndromes of land change in Spain with remote sensing Rodríguez-Rodríguez & Martínez-Vega: Protection of threatened biodiversity in Spain 139

time series, demographic and climatic data. Land Use 2016: risk and resilience in a new era. WWF Interna- Policy 30: 685−702 tional, Gland Turner RK, van der Bergh JCJM, Soderqvist T, Barendregt WWF (World Wildlife Fund) (2017) Preventing paper A, van der Straaten J, Maltby E, van Ierland EC (2000) parks: how to make the EU nature laws work. WWF-EU Ecological-economic analysis of wetlands: scientific inte- Report 2017. http://wwf.panda.org/about_our_ earth/all gration for management and policy. Ecol Econ 35: 7−23 _ publications/ ?291910/ Preventing-Paper-Parks&dm_i= UNESCO (1972) Convention concerning the protection of 2 GI3, 10 RBI,4TLQLQ,2ZEEG,1 (accessed 23 Mar 2017) the world cultural and natural heritage. Adopted by the Zamora-Marín JM, Gutiérrez-Cánovas C, Abellán P, Millán General Conference at its seventeenth session, Paris, 16 A (2016) The role of protected areas in representing November 1972. http: //whc.unesco.org/archive/convent aquatic biodiversity: a test using α, β and γ diversity of ion -en.pdf (accessed 5 Oct 2017) water beetles from the Segura River Basin (SE Spain). WWF (World Wildlife Fund) (2016) Living planet report Limnetica 35: 179−192

Editorial responsibility: Nils Bunnefeld, Submitted: July 24, 2017; Accepted: December 5, 2017 Stirling, UK Proofs received from author(s): February 9, 2018